Information processing methods, programs, and information processing devices.

Abstract

This technology provides the ability to estimate the feel of hitting a golf ball. [Solution] An information processing device 130 according to one embodiment of the present disclosure includes a feature quantity acquisition unit 1303 that acquires feature quantity F based on data of the sound of impact of a golf ball to be estimated for feel, and a feel estimation unit 1305 that estimates the feel of the golf ball to be estimated based on correlation information representing the correlation between the sensory evaluation of the feel by a tester when a test ball is hit and the feature quantity F acquired from the sound of impact at that time, and the feature quantity F acquired by the feature quantity acquisition unit 1303. The feature quantity acquisition unit 1303 acquires feature quantity F based on average sound pressure P1, which is the average sound pressure of the sound of impact of the golf ball to be estimated in the frequency range FA1, and average sound pressure P2, which is the average sound pressure of the sound of impact of the golf ball to be estimated in the frequency range FA2, which includes frequencies lower than the frequency range FA1.

Description

Technical Field

[0001] This disclosure relates to an information processing method and the like.

Background Art

[0002] For example, there is a known technique for estimating the feel of a golf ball without relying on the sensory evaluation of a tester (see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, there is room for improvement in the technique for estimating the feel of a golf ball.

[0005] Therefore, in view of the above problems, an object is to provide a technique capable of estimating the feel of a golf ball.

Means for Solving the Problems

[0006] To achieve the above object, in one embodiment of the present disclosure, an acquisition step in which an information processing device acquires a feature amount based on data of a hitting sound of a golf ball to be estimated for feel; and an estimation step in which the information processing device estimates the feel of the golf ball to be estimated based on information representing the correlation between the sensory evaluation of feel by a tester at the time of hitting a test ball and the feature amount obtained from the hitting sound at the time of hitting, and the feature amount obtained in the acquisition step, In the acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated in a first frequency range, and a second average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated in a second frequency range that includes frequencies lower than the first frequency range. Information processing methods are provided.

[0007] In other embodiments of this disclosure, The information processing device performs a first acquisition step in which it acquires feature quantities based on the sound data of the test ball being hit, The information processing device performs a second acquisition step in which it acquires information on the sensory evaluation of the feel of the test ball when struck by a tester, The information processing device includes a derivation step of deriving information representing the correlation between the feature quantities and the feel of the golf ball, based on the feature quantities and sensory evaluation information obtained in the first acquisition step and the second acquisition step for a plurality of types of test balls, In the first acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a first frequency range, and a second average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a second frequency range that includes frequencies lower than the first frequency range. Information processing methods are provided.

[0008] Furthermore, in yet another embodiment of this disclosure, In an information processing device, An acquisition step to obtain feature quantities based on the sound data of the golf ball being struck, which is the target of the feel estimation, The system performs an estimation step that estimates the feel of the golf ball to be estimated based on information representing the correlation between the sensory evaluation of the feel of the test ball by a tester during impact and the feature quantities obtained from the sound of impact during impact, and the feature quantities obtained in the acquisition step. In the acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated in a first frequency range, and a second average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated in a second frequency range that includes frequencies lower than the first frequency range. A program will be provided.

[0009] Furthermore, in yet another embodiment of this disclosure, In an information processing device, The first acquisition step involves obtaining features based on the sound data of the test ball being hit, A second acquisition step involves obtaining information on the subjective evaluation of the feel of the test ball when struck by a tester. For multiple types of test balls, a derivation step is performed to derive information representing the correlation between the feature quantities and the feel of the golf ball, based on the feature quantities and sensory evaluation information obtained in the first acquisition step and the second acquisition step. In the first acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a first frequency range, and a second average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a second frequency range that includes frequencies lower than the first frequency range. A program will be provided.

[0010] Furthermore, in yet another embodiment of this disclosure, An acquisition unit that acquires feature quantities based on the sound data of the golf ball being struck, which is the target of the feel estimation, The system includes information representing the correlation between the sensory evaluation of the feel of a test ball when struck by a tester and the characteristic quantities obtained from the sound of the impact at that time, and an estimation unit that estimates the feel of the golf ball to be estimated based on the characteristic quantities obtained by the acquisition unit. The acquisition unit acquires the feature amount based on a first average sound pressure that is an average of sound pressures of hitting sounds of the golf ball to be estimated in a first frequency range, and a second average sound pressure that is an average of sound pressures of hitting sounds of the golf ball to be estimated in a second frequency range including frequencies lower than the first frequency range. An information processing apparatus is provided.

[0011] In still another embodiment of the present disclosure, a first acquisition unit that acquires a feature amount based on data of hitting sounds of a test ball, a second acquisition unit that acquires information on a sensory evaluation of hitting feeling by a tester when hitting the test ball, a derivation unit that derives information representing a correlation between the feature amount and the hitting feeling of a golf ball based on the feature amount and the sensory evaluation information acquired by the first acquisition unit and the second acquisition unit for a plurality of types of the test balls, The first acquisition unit acquires the feature amount based on a first average sound pressure that is an average of sound pressures of hitting sounds of the test ball in a first frequency range, and a second average sound pressure that is an average of sound pressures of hitting sounds of the test ball in a second frequency range including frequencies lower than the first frequency range. An information processing apparatus is provided.

Advantages of the Invention

[0012] According to the above-described embodiment, the hitting feeling of a golf ball can be estimated.

Brief Description of the Drawings

[0013] [Figure 1] It is a diagram showing an example of an information processing system. [Figure 2] It is a diagram showing a specific example of a method for measuring a hitting sound of a golf ball by a golf club. [Figure 3] It is a diagram showing a configuration of an example of an information processing apparatus. [Figure 4] It is a functional block diagram showing a configuration of an example of an information processing system. [Figure 5] It is a diagram showing a specific example of data on the hitting sound of a golf ball. [Figure 6] It is a diagram showing a specific example of a method for evaluating the feel of a golf ball. [Figure 7] It is a diagram showing a specific example representing the feature amount of the hitting sound of a test ball and the sensory evaluation of the feel by a tester when hitting the test ball. [Figure 8] It is a diagram for explaining an example of a machine learning method. [Figure 9] It is a diagram for explaining a specific example of information representing the correlation between the feature amount of the hitting sound of a test ball and the sensory evaluation of the feel by a tester when hitting the test ball. [Figure 10] It is a flowchart diagram schematically showing an example of a process related to the derivation of information representing the correlation between the feature amount of the hitting sound of a test ball and the sensory evaluation of the feel by a tester when hitting the test ball. [Figure 11] It is a functional block diagram showing the configuration of an example of an information processing system. [Figure 12] It is a flowchart diagram schematically showing an example of a process related to the estimation of the feel of a golf ball.

Embodiments for Carrying Out the Invention

[0014] Hereinafter, embodiments will be described with reference to the drawings.

[0015] [Information Processing System] Referring to FIGS. 1 to 3, the information processing systems 1 and 2 according to the present embodiment will be described.

[0016] FIG. 1 is a diagram showing an example of the information processing systems 1 and 2. FIG. 2 is a diagram showing a specific example of a method for measuring the hitting sound of a golf ball by a golf club. FIG. 3 is a diagram showing the configuration of the information processing device 30.

[0017] FIG. 2 includes FIGS. 2A and 2B showing an example and other examples of the hitting sound of a golf ball by a golf club.

[0018] As shown in Figure 1, the information processing system 1 includes a microphone 10, a signal processing circuit 20, and an information processing device 30.

[0019] The information processing system 1, in the information processing device 30, derives information for estimating the feel of a golf ball based on the sound of a golf club hitting a golf ball (hereinafter simply referred to as "golf ball impact sound"). The information for estimating the feel of a golf ball based on the golf ball impact sound is information representing the correlation between feature quantities that represent the characteristics of the golf ball impact sound and the sensory evaluation of the feel of the golf ball (hereinafter referred to as "correlation information" for convenience).

[0020] In the following, the golf balls used to derive correlation information may be referred to as "test balls" for convenience.

[0021] Microphone 10 picks up the sound of the test ball being hit.

[0022] For example, as shown in Figure 2A, the tester 40 hits a golf ball 200 (test ball) with a golf club 50. In this example, the microphone 10 is placed near the tester 40's ear. This makes it possible to bring the sound of the impact that the tester 40 actually hears as close as possible to the sound of the impact picked up by the microphone 10.

[0023] Returning to Figure 1, the signal representing the sound data of the test ball being hit, which is collected by the microphone 10, is input to the signal processing circuit 20 via a predetermined communication line such as a one-to-one communication line or a local area network (LAN).

[0024] The signal processing circuit 20 is an electrical circuit that performs predetermined processing on the signal received from the microphone 10. The signal processing circuit 20 includes, for example, an analog-to-digital conversion circuit that converts an analog signal to a digital signal, and a filter circuit that removes noise.

[0025] The signal representing the sound of the test ball being struck, which has been processed by the signal processing circuit 20, is received by the information processing device 30 via a predetermined communication line such as a one-to-one communication line or a local network.

[0026] The information processing device 30 derives correlation information based on the sound data of the test ball's impact, which is acquired from the signal processing circuit 20, and the sensory evaluation information of the feel of the test ball's impact by a tester, which is input separately (hereinafter referred to as "sensory evaluation information").

[0027] The information processing device 30 is connected to the information processing device 130 via a predetermined communication line, such as a local network or a wide area network (WAN) including a mobile communication network, satellite communication network, or internet network. This allows the information processing device 30 to provide the derived correlation information to the information processing device 130, which will be described later.

[0028] The functions of the information processing device 30 can be realized by any hardware or any combination of hardware and software. For example, as shown in Figure 3, the information processing device 30 includes an external interface 31, an auxiliary storage device 32, a memory device 33, a processor 34, a communication interface 36, an input device 37, a display device 38, and an audio output device 39. These components are connected by bus B3.

[0029] The external interface 31 functions as an interface for reading data from and writing data to the recording medium 31A. The recording medium 31A includes, for example, a flexible disk, CD (Compact Disc), DVD (Digital Versatile Disc), BD (Blu-ray® Disc), SD memory card, USB (Universal Serial Bus) memory, etc. This allows the information processing device 30 to read various data used in processing through the recording medium 31A, store it in the auxiliary storage device 32, and install programs that realize various functions.

[0030] Furthermore, the information processing device 30 may acquire various data and programs for processing from external devices through the communication interface 36.

[0031] The auxiliary storage device 32 stores various installed programs, as well as files and data necessary for various processes. The auxiliary storage device 32 includes, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), or flash memory.

[0032] When a program startup command is received, the memory device 33 reads the program from the auxiliary storage device 32 and stores it. The memory device 33 includes, for example, DRAM (Dynamic Random Access Memory) or SRAM (Static Random Access Memory).

[0033] The processor 34 executes various programs loaded from the auxiliary storage device 32 into the memory device 33 and implements various functions related to the information processing device 30 according to the programs. The processor 34 includes, for example, a CPU (Central Processing Unit). The processor may also include a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), etc.

[0034] The communication interface 36 is used as an interface for communicating with external devices. This allows the information processing device 30 to receive signals from the signal processing circuit 20 through the communication interface 36. Furthermore, the information processing device 30 can communicate bidirectionally with, for example, the information processing device 130, through the communication interface 36. The communication interface 36 has multiple types of communication interfaces, for example, to suit the communication method between the connected devices.

[0035] The input device 37 receives various inputs from the user of the information processing device 30.

[0036] The input device 37 includes, for example, an input device that accepts mechanical input from a user (hereinafter referred to as "mechanical input device"). The mechanical input device includes, for example, buttons, toggles, levers, touch panels mounted on the display device 38, touchpads, keyboards, mice, etc., that are provided separately from the display device 38. The input device 37 may also include a voice input device that can accept voice input from a user. The voice input device includes, for example, a microphone that can collect the user's voice. The input device 37 may also include a gesture input device that can accept gesture input from a user. The gesture input device includes, for example, a camera that can capture images of the user's gestures. The input device 37 may also include a biometric input device that can accept biometric input from a user. The biometric input device includes, for example, a camera that can acquire image data containing information about the user's fingerprints or iris.

[0037] The display device 38 visually conveys information to the user by displaying information screens, operation screens, etc. The display device 38 is, for example, a liquid crystal display or an organic EL (electroluminescence) display.

[0038] The sound output device 39 conveys various information to the user audibly by outputting a predetermined sound or voice. The sound output device 39 may be, for example, a buzzer, alarm, speaker, etc.

[0039] Returning to Figure 1, the information processing system 2 includes a microphone 110, a signal processing circuit 120, and an information processing device 130.

[0040] The information processing system 2, using the information processing device 130, estimates the feel of hitting a golf ball based on the sound of the impact.

[0041] Hereafter, the golf ball used for estimating the feel of the ball may be referred to as the "golf ball used for estimation" for convenience.

[0042] Microphone 110 collects the sound of the golf ball being hit, which is the estimated target.

[0043] For example, as shown in Figure 2A, similar to the case of collecting the sound of a test ball being struck, the tester 40 strikes the golf ball 200 (the golf ball to be estimated) with a golf club 50. In this case, the microphone 110 is placed near the ear of the tester 40, similar to the case of microphone 10.

[0044] Furthermore, as shown in Figure 2B, the swing robot 60 may strike the golf ball 200 (the golf ball to be estimated) using the golf club 50. This is because sensory evaluation of the feel of the golf ball to be estimated is unnecessary when estimating the feel of the golf ball to be estimated. This allows for efficient testing of multiple types of golf balls to be estimated. In this case, for example, as shown in Figure 2, the microphone 110 is placed relatively close to the golf ball 200 (the golf ball to be estimated). This ensures that the sound of the golf ball 200 being struck by the swing robot 60 is reliably collected, even under conditions where the sound of the impact is relatively quiet. In addition, the microphone 110 may be placed at a position corresponding to the ear of the tester 40, with the golf ball 200 as the reference point, similar to when sensory evaluation is performed by the tester 40.

[0045] The swing robot 60 can strike a golf ball 200 by, for example, using a driving force such as an electric motor to rotate the arm that holds the golf club 50. Alternatively, the swing robot 60 may be configured in such a way that a person can manually raise the arm to the top position, and the arm rotates downward due to gravity to strike the golf ball 200. In this case, it is possible to prevent the driving sound of the electric motor, etc. from being included in the data collected by the microphone 110. Furthermore, part or all of the surface of the swing robot 60 may be covered with sound-absorbing material as appropriate. This makes it possible to suppress the occurrence of situations where the driving sound of the electric motor, etc. of the swing robot 60 or the operation sound of the arm, etc., are included in the data collected by the microphone 110.

[0046] Returning to Figure 1, the signal representing the estimated sound of the golf ball being struck, collected by the microphone 110, is input to the signal processing circuit 120 via a predetermined communication line, such as a one-to-one communication line or a local network.

[0047] The signal processing circuit 120 has the same function as the signal processing circuit 20 and is an electrical circuit that performs predetermined processing on the signal picked up by the microphone 110.

[0048] The signal representing the estimated sound of the golf ball being struck, which has been processed by the signal processing circuit 120, is received by the information processing device 130 via a predetermined communication line such as a one-to-one communication line or a local network.

[0049] The information processing device 130 estimates the feel of the golf ball being hit based on the sound data of the golf ball being hit, which is acquired from the signal processing circuit 120, and the correlation information provided by the information processing device 30.

[0050] The functions of the information processing device 130 can be realized by any hardware or any combination of hardware and software. For example, the information processing device 130 has the same hardware components as the information processing device 30 described above (see Figure 3).

[0051] In the following, when referring to the hardware components of the information processing device 130, components with the same function as those of the information processing device 30 may, for convenience, be described by adding a "1" to the beginning of the code of the components of the information processing device 30. For example, the display device as a component of the information processing device 130 may be referred to as "display device 138," by referencing the display device 38 of the information processing device 30.

[0052] [Method for deriving correlation information] Refer to Figures 4 to 9 to specifically explain how correlation information is derived.

[0053] Figure 4 is a functional block diagram showing the configuration of an example of information processing system 1. Figure 5 is a diagram showing a specific example of data on the sound of a golf ball being struck. Figure 6 is a diagram showing a specific example of a method for evaluating the feel of a golf ball being struck. Figure 7 is a diagram showing a specific example of the characteristic quantities of the sound of a test ball being struck, and the sensory evaluation of the feel by a tester when the test ball is struck. Figure 8 is a diagram illustrating an example of a machine learning method. Figure 9 is a diagram illustrating a specific example of information showing the correlation between the characteristic quantities of the sound of a test ball being struck and the sensory evaluation of the feel by a tester when the test ball is struck.

[0054] Figure 5 includes Figure 5A, which shows the time-series sound pressure data D1 of the sound of the test ball being struck, acquired from the signal processing circuit 20, and Figure 5B, which shows the frequency domain sound pressure data D2 obtained as a result of frequency analysis.

[0055] As shown in Figure 4, the information processing device 30 includes, as functional units, a sound impact data acquisition unit 301, a sound impact data storage unit 301A, a frequency analysis unit 302, a feature quantity acquisition unit 303, a feature quantity information storage unit 303A, a sensory evaluation information acquisition unit 304, a sensory evaluation information storage unit 304A, a data generation unit 305, a data storage unit 305A, a correlation information derivation unit 306, a correlation information storage unit 306A, and a correlation information distribution unit 307. These functions are realized, for example, by loading a program installed in the auxiliary storage device 32 into the memory device 33 and executing it in the processor 34. In addition, the functions of various storage units are realized, for example, by storage areas defined in the auxiliary storage device 32, etc.

[0056] The impact sound data acquisition unit 301 acquires the impact sound data of the test ball, which is input from the signal processing circuit 20 to the information processing device 30, and registers it in the impact sound data storage unit 301A.

[0057] The impact sound data storage unit 301A stores data of the impact sound of the test balls, associated with information that identifies each individual test ball (identification information).

[0058] The frequency analysis unit 302 performs frequency analysis on the test ball impact sound data. This allows the frequency analysis unit 302 to obtain the frequency spectrum data of the test ball impact sound data, which is time-series data. For example, as shown in Figure 5A, the frequency analysis unit 302 uses, for example, FFT (Fast Fourier Transform) analysis to convert the time-series sound pressure data D1 of the test ball impact sound into frequency-domain sound pressure data D2. The frequency-domain sound pressure data D2 is the frequency spectrum data of the time-series sound pressure data D1.

[0059] The feature acquisition unit 303 acquires a feature F representing the characteristics of the sound of the test ball being hit, based on the frequency spectrum data of the sound of the test ball being hit, which is output by the frequency analysis unit 302, and registers it in the feature information storage unit 303A.

[0060] Feature quantity F is obtained based on the average sound pressure of the impact sound in frequency range FA1 (hereinafter referred to as "average sound pressure") P1 and the average sound pressure of the impact sound in frequency range FA2, which is different from frequency range FA1 (average sound pressure) P2. For example, feature quantity F is the ratio of average sound pressure P1 to average sound pressure P2, specifically the value obtained by dividing average sound pressure P2 by average sound pressure P1 (F = P2 / P1). Furthermore, the numerator and denominator of the ratio between average sound pressures P1 and P2 may be reversed in feature quantity F, or it may be a value obtained by multiplying the ratio of average sound pressure P1 to average sound pressure P2 by a coefficient.

[0061] The frequency range FA2 includes a frequency range lower than the frequency range FA1. For example, as shown in Figure 5B, the frequency range FA2 is set to a range that partially or completely overlaps with the frequency range FA1 and includes a frequency range lower than the frequency range FA1. Alternatively, the frequency range FA2 may be a frequency range lower than the frequency range FA1 that does not overlap with the frequency range FA1.

[0062] For example, as shown in Figure 5B, frequency range FA1 is the frequency range where the lower limit is frequency f1 and the upper limit is frequency f2 (>0), and frequency range FA2 is the frequency range where the lower limit is frequency f0 (not shown) which is lower than frequency f1 and the upper limit is frequency f2.

[0063] For example, frequency f1 is 5000 Hz (Hertz), and frequency f2 is 20000 Hz (Hertz). Also, frequency f1 may be 5000 Hz higher, and frequency f2 may be lower than 20000 Hz. For example, frequency f0 is 0 Hz (Hertz). Also, frequency f0 may be higher than 0 Hz.

[0064] Frequency range FA1 is the frequency range in which relatively high-pitched sounds of the golf ball impact are present, compared to frequency range FA2. Frequency range FA2 is the frequency range in which relatively low-pitched sounds of the golf ball impact are present, compared to frequency range FA1. Furthermore, if frequency range FA2 partially or completely overlaps with frequency range FA1, it is the frequency range in which a relatively wide frequency range of the golf ball impact sound, including relatively low-pitched sounds, is present, compared to frequency range FA1. Therefore, by using the average sound pressure P1 of frequency range FA1 and the average sound pressure P2 of frequency range FA2, the feature acquisition unit 303 can acquire an appropriate feature quantity F that represents the characteristics of the golf ball impact sound.

[0065] The feature information storage unit 303A stores the feature quantities F acquired by the feature quantity acquisition unit 303 in a manner that is associated with information that identifies each individual test ball (identification information).

[0066] The sensory evaluation information acquisition unit 304 acquires information (sensory evaluation information) that represents the sensory evaluation of the feel of hitting a test ball by a tester, which is input from an external source, and registers it in the sensory evaluation information storage unit 304A. For example, the sensory evaluation information is manually input into the information processing device 30 by the tester or an operator via the input device 37. Alternatively, the sensory evaluation information may be input into the information processing device 30 via the external interface 31 or the communication interface 36 as information that has already been compiled by another device.

[0067] The evaluation criteria (also called "evaluation axes") for the feel of a golf ball may be one type, two types, or three or more types.

[0068] For example, as shown in Figure 6, two metrics, "hardness" and "weight," are used to evaluate the feel of a golf ball.

[0069] "Hardness" is an evaluation index that represents whether the golf ball felt hard or soft to the tester when struck. "Weight" is an evaluation index that represents whether the golf ball felt heavy or light to the tester when struck.

[0070] For example, the evaluation indices for "hardness" and "weight" are both quantified numerically on a scale from "+5" in the positive direction to "-5" in the negative direction, with "0" (zero) as the baseline. Sensory evaluation is performed on an 11-point scale, represented by 5 positive integer levels up to "+5" and 5 negative integer levels up to "-5," with "0" as the central point. The "0" level represents the average feel by the tester. For example, if there is a reference golf ball, the "0" level corresponds to the feel of that golf ball. In this case, the tester may test the actual test ball after testing the reference golf ball.

[0071] For example, relatively high-frequency sounds tend to give listeners a relatively hard impression regarding the "hardness" evaluation index. Also, relatively loud relatively high-frequency sounds tend to give listeners a relatively heavy impression regarding the "weight" evaluation index, and vice versa. Therefore, by combining the average sound pressure P2 of the frequency range FA2, which represents the characteristics of relatively high-frequency sounds in the sound of a golf ball being struck, with the average sound pressure P1 of the frequency range FA1, which represents the characteristics of relatively high-frequency sounds in the sound of a golf ball being struck, the feature quantity F is expected to have a relatively high correlation with the sensory evaluation of the "hardness" and "weight" evaluation indices.

[0072] Returning to Figure 4, the sensory evaluation information storage unit 304A stores the sensory evaluation information acquired by the sensory evaluation information acquisition unit 304 in a form associated with information that identifies each individual test ball (identification information).

[0073] The data generation unit 305 generates data for the correlation information derivation unit 306 to derive correlation information and registers it in the data storage unit 305A.

[0074] For example, the data generation unit 305 generates training data for the correlation information derivation unit 306 to generate correlation information (i.e., a trained model) using machine learning (specifically, supervised learning). Based on the information registered in the feature information storage unit 303A and the sensory evaluation information storage unit 304A, the data generation unit 305 generates training data for each test ball, which is a combination of feature F as input and the sensory evaluation of the feel when hit by a tester as the correct output. For example, as shown in Figure 7, training data is generated for multiple test balls "A", "B", "C", "D", etc., based on a combination of feature F and the sensory evaluation of the feel when hit. In this way, the data generation unit 305 can generate a training dataset as a group of training data for multiple types of test balls.

[0075] Returning to Figure 4, the correlation information derivation unit 306 uses the data stored in the data storage unit 305A to derive correlation information and registers it in the correlation information storage unit 306A.

[0076] For example, the correlation information derivation unit 306 performs supervised learning of the base learning model based on the training dataset stored in the data storage unit 305A, and derives correlation information as a trained model. If there are multiple types of evaluation metrics (evaluation axes) for feel, the correlation information derivation unit 306 performs machine learning of the base learning model for each type of evaluation metric for feel, and derives correlation information for each type of evaluation metric for feel.

[0077] For example, as shown in Figure 8, when input data 71 ("X") from the training data is input to the input layer of the learning model 72 ("M"), the learning model 72 outputs output data 73 ("Y") corresponding to the input data 71 from its output layer. In supervised learning, the output data 73 of the learning model 72 and the ground truth data 74 ("T") from the training data are given to the loss function 75 to obtain the difference amount 76 ("L") between the output data 73 of the learning model 72 and the ground truth data 74. The learning model 72 is optimized by updating the internal coefficients and weights of the learning model 72 for multiple training data included in the training dataset, and as a result, the correlation information derivation unit 306 can obtain correlation information as a trained model.

[0078] The learning model 72 can be expressed, for example, as a formula for the linear correlation (first-order correlation) between the feature quantity F and the sensory evaluation of the feel of the touch, as shown in equation (1) below.

[0079] (Feel) = (Coefficient α) × (Feature F) + (Constant β) ... (1)

[0080] In this case, machine learning corresponds to obtaining a first-order regression equation through so-called regression analysis. As a result, for example, as shown in Figure 9, the correlation information derivation unit 306 can derive a linear correlation formula (first-order regression equation) that represents the correlation between the feature quantity F and the sensory evaluation of the feel of the ball, as a trained model (see dotted line in the figure).

[0081] Furthermore, the regression equation does not have to be a first-order regression equation like equation (1); it may be a second-order or higher regression equation.

[0082] Furthermore, the learning model 72 may use a neural network such as a DNN (Deep Neural Network). In this case, supervised learning of the learning model 72 is performed using a learning algorithm such as backpropagation.

[0083] Returning to Figure 4, the correlation information derived by the correlation information derivation unit 306 is stored in the correlation information storage unit 306A.

[0084] Furthermore, after the correlation information has been derived, information on feature quantities F and sensory evaluation information for other types of test balls may be additionally registered in the feature quantity information storage unit 303A and the sensory evaluation information storage unit 304A. In this case, the data generation unit 305 may add training data, and the correlation information derivation unit 306 may use the additional training data to perform additional training on the trained model or retrain the trained model 72, thereby updating the correlation information in the correlation information storage unit 306A.

[0085] When the correlation information in the correlation information storage unit 306A is updated, the correlation information may be updated by overwriting the old version of the correlation information, or the correlation information may be updated by retaining the old version of the correlation information and newly registering the new version of the correlation information. In the latter case, for example, if there is a defect in the new version of the correlation information, it is possible to revert to the old version of the correlation information.

[0086] The correlation information distribution unit 307 distributes the correlation information stored in the correlation information storage unit 306A to the information processing device 130.

[0087] For example, the correlation information distribution unit 307 distributes correlation information to the information processing device 130 in response to a signal requesting the distribution of correlation information received from the information processing device 130 via the communication interface 36. Alternatively, the correlation information distribution unit 307 may distribute correlation information to the information processing device 130 at a predetermined timing, regardless of a request from the information processing device 130, when new correlation information is registered in the correlation information storage unit 306A, or when the correlation information in the correlation information storage unit 306A is updated.

[0088] [Processing related to the derivation of correlation information] Refer to Figure 10 to explain a specific example of the process for deriving correlation information.

[0089] Figure 10 is a flowchart illustrating an example of the process for deriving information (correlation information) that represents the correlation between the characteristic quantities of the sound of impact of a test ball and the sensory evaluation of the feel of the ball by a tester during impact.

[0090] As shown in Figure 10, in step S102, the impact sound data acquisition unit 301 acquires the impact sound data of the test ball, which is taken in from the signal processing circuit 20 in accordance with the test ball being hit by the tester, and registers it in the impact sound data storage unit 301A.

[0091] After the processing in step S102, the processing in step S104 is performed.

[0092] In step S104, the frequency analysis unit 302 performs frequency analysis of the impact sound data for each test ball stored in the impact sound data storage unit 301A, and outputs frequency spectrum data.

[0093] After the processing in step S104, the processing in step S106 is performed.

[0094] In step S106, the feature acquisition unit 303 acquires a feature F representing the characteristics of the sound of the test ball being hit, based on the frequency spectrum data of the sound of the test ball being hit output from the frequency analysis unit 302, and registers it in the feature information storage unit 303A.

[0095] After the processing in step S106, the processing in step S108 is performed.

[0096] In step S108, the sensory evaluation information acquisition unit 304 acquires information (sensory evaluation information) representing the sensory evaluation of the feel of each test ball by a tester, which is input, for example, through the input device 37, and registers it in the sensory evaluation information storage unit 304A.

[0097] After the processing in step S108, the processing in step S110 is performed.

[0098] Note that the process in step S108 may be performed as a separate process from the series of processes in this flowchart. In this case, the process in step S110 will be performed after the process in step S106.

[0099] In step S110, the data generation unit 305 generates training data for deriving correlation information based on the feature quantity F information and sensory evaluation information for each test ball, which are registered in the feature quantity information storage unit 303A and the sensory evaluation information storage unit 304A, and registers it in the data storage unit 305A.

[0100] After the processing in step S110, the processing in step S112 is performed.

[0101] In step S112, the correlation information derivation unit 306 derives correlation information by performing machine learning (supervised learning) based on a training dataset of multiple training data stored in the data storage unit 305A, and registers it in the correlation information storage unit 306A.

[0102] After the processing in step S112, the processing in step S114 is performed.

[0103] In step S114, the correlation information distribution unit 307 distributes the latest correlation information stored in the correlation information storage unit 306A to the information processing device 130 via the communication interface 36.

[0104] After the processing in step S114, the processing in this flowchart is completed.

[0105] [Method for estimating the feel of a golf ball] Referring to Figure 11, we will now specifically explain the method for estimating the feel of the golf ball being tested.

[0106] Figure 11 is a functional block diagram showing an example configuration of information processing system 2.

[0107] As shown in Figure 11, the information processing device 130 includes, as functional units, a sound impact data acquisition unit 1301, a sound impact data storage unit 1301A, a frequency analysis unit 1302, a feature quantity acquisition unit 1303, a correlation information storage unit 1304, a feel-inducing unit 1305, and an estimation result output unit 1306.

[0108] The impact sound data acquisition unit 1301 acquires the impact sound data of the golf ball to be estimated, which is input from the signal processing circuit 120 to the information processing device 130, and registers it in the impact sound data storage unit 1301A.

[0109] The frequency analysis unit 1302 performs frequency analysis on the data of the impact sound of the golf ball to be estimated, which is stored in the impact sound data storage unit 1301A. The frequency analysis unit 1302 can output frequency spectrum data of the impact sound data of the golf ball to be estimated in the same manner as the frequency analysis unit 302 described above.

[0110] The feature acquisition unit 1303 acquires feature quantities F that represent the characteristics of the impact sound of the golf ball to be estimated, based on the frequency spectrum data of the impact sound data of the golf ball to be estimated, which is output by the frequency analysis unit 1302. The feature acquisition unit 1303 can acquire the feature quantities F of the golf ball to be estimated in the same manner as the feature acquisition unit 303 described above.

[0111] The correlation information storage unit 1304 stores correlation information distributed from the information processing device 30 and received via the communication interface 136. If there are multiple types of evaluation indicators for feel, the correlation information storage unit 1304 stores correlation information for each type of evaluation indicator for feel.

[0112] The feel estimation unit 1305 estimates the feel of the golf ball based on the feature quantities F of the golf ball to be estimated, which are acquired by the feature quantity acquisition unit 1303, and the correlation information stored in the correlation information storage unit 1304. For example, the feel estimation unit 1305 can obtain an estimated level of the feel of the golf ball as output data from the correlation information by providing the feature quantities F of the golf ball to be estimated as input data to the correlation information, which is a trained model. If there are multiple types of feel evaluation metrics, the feel estimation unit 1305 obtains an estimated result of the feel of the golf ball for each evaluation metric based on the feature quantities F of the golf ball to be estimated and the correlation information for each feel evaluation metric.

[0113] The estimation result output unit 1306 outputs the estimation result of the feel of the golf ball being estimated, which is output from the feel estimation unit 1305, to the user of the information processing device 130 via the display device 138 and the sound output device 139.

[0114] This allows the user of the information processing device 130 to confirm the estimated result of the feel of the golf ball being hit.

[0115] [Processing related to estimating the feel of hitting a golf ball] Referring to Figure 11, a specific example of the process for estimating the feel of the golf ball being hit will be explained.

[0116] Figure 11 is a flowchart illustrating an example of the process for estimating the feel of hitting a golf ball.

[0117] As shown in Figure 11, in step S202, the impact sound data acquisition unit 1301 acquires the impact sound data of the golf ball to be estimated, which is taken in from the signal processing circuit 120 in accordance with the test swing of the golf ball to be estimated, and registers it in the impact sound data storage unit 1301A.

[0118] After the processing in step S202, the processing in step S204 is performed.

[0119] In step S204, the frequency analysis unit 1302 performs frequency analysis on the data of the impact sound of the golf ball to be estimated, which is stored in the impact sound data storage unit 1301A, and outputs frequency spectrum data.

[0120] After the processing in step S204, the processing in step S206 is performed.

[0121] In step S206, the feature acquisition unit 1303 acquires feature quantities F that represent the characteristics of the golf ball to be estimated, based on the frequency spectrum data of the sound of the golf ball being hit, which is output from the frequency analysis unit 1302.

[0122] After the processing in step S206, the processing in step S208 is performed.

[0123] In step S208, the feel estimation unit 1305 estimates the feel of the golf ball based on the feature quantity F of the golf ball to be estimated, which was acquired by the feature quantity acquisition unit 1303, and the correlation information that has been previously registered in the correlation information storage unit 1304.

[0124] After the processing in step S208, the processing in step S210 is performed.

[0125] In step S210, the estimation result output unit 1306 outputs the estimation result of the feel of the golf ball to be estimated, which is output from the feel estimation unit 1305, to the user of the information processing device 130 via the display device 138.

[0126] After the processing in step S210, the processing in this flowchart is completed.

[0127] [Other embodiments] Other embodiments will be described.

[0128] The embodiments described above may be modified or altered as appropriate. Hereinafter, examples of modifications or alterations to the embodiments described above will be referred to as "modified versions" for convenience.

[0129] For example, in the embodiment described above, the functions of the signal processing circuit 20 may be built into the microphone 10.

[0130] Furthermore, in the embodiments and their modifications described above, the function of the signal processing circuit 120 may be built into the microphone 110, similar to the case of the signal processing circuit 20.

[0131] Furthermore, in the embodiments and variations thereof described above, the functions of the information processing device 30 may be implemented in a distributed manner by multiple devices. For example, in addition to the information processing device 30, an external storage device may be provided for storing various types of information to which the functions of the internal storage area of ​​the information processing device 30 can be transferred. Also, various processes for realizing the functions of the information processing device 30 may be implemented in a distributed manner by multiple information processing devices.

[0132] Furthermore, in the embodiments and modifications thereof described above, the functions of the information processing device 130 may be implemented in a distributed manner by multiple devices, similar to the case of the information processing device 30.

[0133] Furthermore, in the embodiments and their modifications described above, the functions of information processing system 2 are integrated into information processing system 1, and information processing system 2 may be omitted. In this case, the functions of microphone 110 are integrated into microphone 10, the functions of signal processing circuit 120 are integrated into signal processing circuit 20, and the functions of information processing device 130 are integrated into information processing device 30.

[0134] Furthermore, in the embodiments and modifications thereof described above, the information processing device 130 may estimate the feel of the golf ball to be estimated based on a plurality of types of feature quantities, including feature quantity F. In this case, the information processing device 30 acquires correlation information representing the correlation between a plurality of types of feature quantities of the sound of impact of the test ball and the sensory evaluation of the feel of impact by a tester when the test ball is struck. For example, additional feature quantities representing the characteristics of the sound of impact from a golf ball as a sound source, as disclosed in the aforementioned Patent Document 1 by the same inventor, may be adopted. In this case, as disclosed in Patent Document 1, the natural frequency of the golf ball is obtained by measuring the vibration characteristics of the golf ball, and feature quantities are obtained from the peaks and average values ​​of the sound pressure in the frequency band near the natural frequency in the frequency spectrum of the golf ball impact sound data.

[0135] [Effect] The operation of the information processing method, program, and information processing device according to this embodiment will be described.

[0136] In a first aspect of this embodiment, an information processing method is provided that includes an acquisition step and an estimation step. The acquisition step is, for example, step S206 described above. The estimation step is, for example, step S208 described above. Specifically, in the acquisition step, the information processing device acquires feature quantities based on data of the sound of impact of a golf ball whose feel is to be estimated. In the estimation step, the information processing device estimates the feel of the golf ball to be estimated based on information representing the correlation between the sensory evaluation of the feel by a tester when the test ball is hit and the feature quantities acquired from the sound of impact at that time, and the feature quantities acquired in the acquisition step. The information processing device is, for example, the information processing device 130 described above. In the acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the sound of impact of the golf ball to be estimated in a first frequency range, and a second average sound pressure, which is the average sound pressure of the sound of impact of the golf ball to be estimated in a second frequency range that includes frequencies lower than the first frequency range. The first frequency range is, for example, the frequency range FA1 described above. The first average sound pressure is, for example, the average sound pressure P1. The second frequency range is, for example, the frequency range FA2 described above. The second average sound pressure is, for example, the average sound pressure P2.

[0137] Furthermore, in the first aspect of this embodiment, a program may be provided that causes the information processing device to perform the acquisition step and the estimation step.

[0138] Furthermore, in the first aspect of this embodiment, an information processing device comprising an acquisition unit and an estimation unit may be provided. The acquisition unit is, for example, the feature quantity acquisition unit 1303 described above. The estimation unit is, for example, the feel estimation unit 1305 described above. Specifically, the acquisition unit acquires feature quantities based on data of the sound of impact of a golf ball whose feel is to be estimated. The estimation unit estimates the feel of the golf ball to be estimated based on information representing the correlation between the sensory evaluation of the feel by a tester when the test ball is hit and the feature quantities acquired from the sound of impact at that time, and the feature quantities acquired by the acquisition unit. The acquisition unit then acquires the feature quantities based on a first average sound pressure, which is the average sound pressure of the sound of impact of the golf ball to be estimated in a first frequency range, and a second average sound pressure, which is the average sound pressure of the sound of impact of the golf ball to be estimated in a second frequency range that includes frequencies lower than the first frequency range.

[0139] As a result, in this embodiment, the information processing device can estimate the feel of the golf ball being estimated without relying on sensory evaluation.

[0140] In particular, when using a putter club, where the impact sound is quiet and it is difficult to distinguish the sound originating from the golf ball, it can be difficult to extract characteristic features of the impact sound originating from the golf ball, taking into account the natural frequency of the golf ball, etc.

[0141] In contrast, in this embodiment, the information processing device does not need to distinguish the impact sound originating from a golf ball, and can appropriately estimate the feel of the golf ball being hit, even when the impact sound is quiet, such as when using a putter club.

[0142] Furthermore, the first frequency range represents the characteristics of relatively high-pitched sounds among the impact sounds of the golf ball being estimated, and the second frequency range includes lower frequencies and represents the characteristics of sounds including relatively low-pitched sounds among the impact sounds. Therefore, in this embodiment, the information processing device can obtain feature quantities that appropriately represent the characteristics of the impact sound of the golf ball being estimated by using the first average sound pressure of the first frequency range and the second average sound pressure of the second frequency range among the impact sounds of the golf ball being estimated. Thus, the information processing device can estimate the feel of the golf ball being hit more appropriately.

[0143] Furthermore, in a second aspect of this embodiment, based on the first aspect described above, the feature quantity may be the ratio of the first average sound pressure and the second average sound pressure.

[0144] As a result, in this embodiment, the information processing device can acquire feature quantities that represent the characteristics of the sound of impact of the golf ball being estimated.

[0145] Furthermore, in a third aspect of this embodiment, based on the first or second aspect described above, the first frequency range may be defined within the range of 5000Hz to 20000Hz, and the second frequency range may be defined within the range of 0Hz to 20000Hz.

[0146] As a result, in this embodiment, the information processing device can acquire feature quantities that represent the characteristics of the sound of impact of the golf ball being estimated.

[0147] Furthermore, in a fourth aspect of this embodiment, assuming any one of the first to third aspects described above, the first frequency range and the second frequency range may include two or more natural frequencies of the golf ball being estimated.

[0148] As a result, in this embodiment, the information processing device can appropriately estimate the feel of the golf ball being hit without considering the natural frequency of the golf ball being estimated.

[0149] Furthermore, in a fifth aspect of this embodiment, assuming any one of the first to fourth aspects described above, the information may be a mathematical formula representing the correlation between the feature quantity and the level of feel of the golf ball.

[0150] As a result, in this embodiment, the information processing device can estimate the level of the feel of the golf ball's impact using a mathematical formula based on the characteristic quantities of the impact sound of the golf ball being estimated.

[0151] Furthermore, in a sixth aspect of this embodiment, an information processing method is provided that includes a first acquisition step, a second acquisition step, and a derivation step. The first acquisition step is, for example, step S106 described above. The second acquisition step is, for example, step S108 described above. The derivation step is, for example, step S112 described above. Specifically, in the first acquisition step, the information processing device acquires feature quantities based on data of the sound of impact of the test ball. In the second acquisition step, the information processing device acquires information of the sensory evaluation of the feel of the test ball by a tester when it is hit. In the derivation step, the information processing device derives information representing the correlation between the feature quantities and the feel of the golf ball, based on the feature quantities and the sensory evaluation information acquired in the first and second acquisition steps for a plurality of types of test balls. The information processing device is, for example, the information processing device 30 described above. Then, in the first acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a first frequency range, and a second average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a second frequency range that includes frequencies lower than the first frequency range.

[0152] Furthermore, in a sixth aspect of this embodiment, a program is provided that causes the information processing device to perform the first acquisition step, the second acquisition step, and the derivation step.

[0153] Furthermore, in a sixth aspect of this embodiment, an information processing device is provided that includes a first acquisition unit, a second acquisition unit, and a derivation unit. The first acquisition unit is, for example, the feature quantity acquisition unit 303 described above. The second acquisition unit is, for example, the sensory evaluation information acquisition unit 304 described above. The derivation unit is, for example, the correlation information derivation unit 306 described above. Specifically, the first acquisition unit acquires feature quantities based on data of the sound of impact of a test ball. The second acquisition unit acquires information on the sensory evaluation of the feel of the test ball by a tester when it is hit. The derivation unit derives information representing the correlation between the feature quantities and the feel of the golf ball, based on the feature quantities and sensory evaluation information acquired by the first and second acquisition units for a plurality of types of test balls. The first acquisition unit then acquires the feature quantities based on a first average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a first frequency range, and a second average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a second frequency range that includes frequencies lower than the first frequency range.

[0154] In this embodiment, the information processing device can acquire information representing the correlation between the characteristic sound of the test ball being struck and the sensory evaluation of the feel of the test ball by a tester at the time of impact, for estimating the feel of the golf ball being struck.

[0155] Furthermore, in this embodiment, the information processing device can acquire information for estimating the feel of a golf ball based on features that do not require the identification of impact sounds originating from a golf ball. Therefore, the correlation between the features acquired based on the first average sound pressure in a first frequency range and the second average sound pressure in a second frequency range of the impact sounds of the test ball and the sensory evaluation of the feel is relatively high. Thus, in this embodiment, the information processing device can acquire information for more appropriately estimating the feel of a golf ball.

[0156] Although embodiments have been described in detail above, this disclosure is not limited to these specific embodiments, and various modifications and changes are possible within the scope of the gist described in the claims. [Explanation of symbols]

[0157] 1. Information Processing System 2. Information Processing Systems 10 Microphones 20 Signal Processing Circuits 30 Information Processing Devices 110 Microphone 120 Signal Processing Circuits 130 Information Processing Devices 301 Impact sound data acquisition unit 301A Impact sound data storage unit 302 Frequency Analysis Unit 303 Feature Acquisition Unit 303A Feature Information Storage Unit 304 Sensory Evaluation Information Acquisition Department 304A Sensory Evaluation Information Storage Unit 305 Data Generation Unit 305A Data Storage Unit 306 Correlation Information Derivation Unit 306A Correlation Information Storage Unit 307 Correlation Information Distribution Department 1301 Impact sound data acquisition unit 1301A Impact sound data storage unit 1302 Frequency Analysis Unit 1303 Feature acquisition unit 1304 Correlation Information Storage Unit 1305 Impact feel estimation unit 1306 Estimated Result Output Unit F-Features f0 frequency f1 frequency f2 frequency FA1 frequency range FA2 frequency range P1 Average sound pressure P2 Average sound pressure

Claims

1. The information processing device acquires feature quantities based on the sound data of the golf ball being struck, which is the target of the feel estimation. The information processing device includes information representing the correlation between the sensory evaluation of the feel of the test ball when struck by a tester and the feature quantities obtained from the sound of the impact at that time, and an estimation step of estimating the feel of the golf ball to be estimated based on the feature quantities obtained in the acquisition step. In the acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated in a first frequency range, and a second average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated in a second frequency range that includes frequencies lower than the first frequency range. Information processing methods.

2. The aforementioned feature quantity is the ratio of the first average sound pressure to the second average sound pressure. The information processing method according to claim 1.

3. The aforementioned first frequency range is defined within the range of 5000 Hz to 20000 Hz. The aforementioned second frequency range is defined within the range from 0 Hz to 20,000 Hz. The information processing method according to claim 1 or 2.

4. The first frequency range and the second frequency range include two or more natural frequencies of the golf ball being estimated. The information processing method according to claim 1 or 2.

5. The aforementioned information is a mathematical formula that represents the correlation between the aforementioned feature quantity and the level of feel of the golf ball. The information processing method according to claim 1 or 2.

6. The information processing device performs a first acquisition step in which it acquires feature quantities based on the sound data of the test ball being hit, The information processing device performs a second acquisition step in which it acquires information on the sensory evaluation of the feel of the test ball when struck by a tester, The information processing device includes a derivation step of deriving information representing the correlation between the feature quantities and the feel of the golf ball, based on the feature quantities and sensory evaluation information obtained in the first acquisition step and the second acquisition step for a plurality of types of test balls, In the first acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a first frequency range, and a second average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a second frequency range that includes frequencies lower than the first frequency range. Information processing methods.

7. In an information processing device, An acquisition step to obtain feature quantities based on the sound data of the golf ball being struck, which is the target of the feel estimation, The system performs an estimation step that estimates the feel of the golf ball to be estimated based on information representing the correlation between the sensory evaluation of the feel of the test ball by a tester during impact and the feature quantities obtained from the sound of impact during impact, and the feature quantities obtained in the acquisition step. In the acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated in a first frequency range, and a second average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated in a second frequency range that includes frequencies lower than the first frequency range. program.

8. In an information processing device, The first acquisition step involves obtaining feature quantities based on the sound data of the test ball being hit, A second acquisition step involves obtaining information on the subjective evaluation of the feel of the test ball when struck by a tester. For multiple types of test balls, a derivation step is performed to derive information representing the correlation between the feature quantities and the feel of the golf ball, based on the feature quantities and sensory evaluation information obtained in the first acquisition step and the second acquisition step. In the first acquisition step, the feature quantities are acquired based on a first average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a first frequency range, and a second average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a second frequency range that includes frequencies lower than the first frequency range. program.

9. An acquisition unit that acquires feature quantities based on the sound data of the golf ball being struck, which is the target of the feel estimation, The system includes information representing the correlation between the sensory evaluation of the feel of a test ball when struck by a tester and the characteristic quantities obtained from the sound of the impact at that time, and an estimation unit that estimates the feel of the golf ball to be estimated based on the characteristic quantities obtained by the acquisition unit. The acquisition unit acquires the feature quantities based on a first average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated within a first frequency range, and a second average sound pressure, which is the average sound pressure of the impact sound of the golf ball to be estimated within a second frequency range that includes frequencies lower than the first frequency range. Information processing device.

10. A first acquisition unit acquires feature quantities based on the sound data of the test ball being hit, A second acquisition unit that acquires information on the sensory evaluation of the feel of the test ball when struck by a tester, The system includes a derivation unit that, for multiple types of test balls, derives information representing the correlation between the feature quantities and the feel of the golf ball, based on the feature quantities and sensory evaluation information obtained by the first acquisition unit and the second acquisition unit, The first acquisition unit acquires the feature quantities based on a first average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a first frequency range, and a second average sound pressure, which is the average sound pressure of the sound of the test ball being struck in a second frequency range that includes frequencies lower than the first frequency range. Information processing device.

Images